A combined ligand- and structure-based virtual screening protocol identifies submicromolar PPARγ partial agonists

Peroxisome proliferator-activated receptor γ (PPARγ) is involved in expression of genes that control glucose and lipid metabolism. PPARγ is the molecular target of the thiazolidinedione (TZD) class of antidiabetic drugs. However, despite their clinical use these drugs are associated with numerous adverse effects, which are related to their full activation of PPARγ transcriptional responses. PPARγ partial agonists are the focus of development efforts towards second-generation PPARγ modulators with favorable pharmacology, potent insulin sensitization without the severe full agonists' adverse effects. In order to identify novel PPARγ partial agonist lead compounds, we developed a virtual screening protocol based on three-dimensional ligand-shape similarity and docking. Prioritization gave 235 compounds for experimental screening from the National Institutes of Health (NIH) Molecular Libraries Small Molecule Repository (MLSMR)-a chemical library containing 340 000 compounds. Seven novel potent partial agonists were confirmed in cell-based transactivation and competitive binding assays. Our results illustrate a well-designed virtual screening campaign successfully identifying novel lead compounds as potential entry points for the development of antidiabetic drugs.